Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 270 - Bayesian Data Science and Analytics
Type: Topic Contributed
Date/Time: Tuesday, August 4, 2020 : 1:00 PM to 2:50 PM
Sponsor: Section on Bayesian Statistical Science
Abstract #313789
Title: Bayesian Seemingly Unrelated Mean and Quantile Stochastic Frontier Model for Demand
Author(s): Ehsan S Soofi* and Mahsa Mardikoraem
Companies: University of Wisconsin-Milwaukee and Unversity of Wisconsin-Milwaukee
Keywords:
Abstract:

Suppliers of a product receive information about the quantiles of demand from retailers and about the mean demand from consumers. The supplier may or may not be able to meet the demand due to the limited inventory which leads to the demand distribution to stochastically dominate the sales distribution. This relationship allows using the stochastic frontier models, to estimate the mean and the quantile of the demand based on data on sales and covariates. The interrelationship between the mean and quantiles of a distribution induces dependence between the regression equations for the mean and the quantile. A Bayesian formulation for estimating the model parameters and developing predictive models is proposed. The efficacy of this model is illustrated.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2020 program